RE: efficacy of MODULE_LICENSE

From: Thomas Gleixner <tglx@linutronix.de>
To: "Zavras, Alexios" <alexios.zavras@intel.com>
Cc: J Lovejoy <opensource@jilayne.com>,
	"linux-spdx@vger.kernel.org" <linux-spdx@vger.kernel.org>
Subject: RE: efficacy of MODULE_LICENSE
Date: Wed, 10 Jul 2019 20:55:58 +0200 (CEST)	[thread overview]
Message-ID: <alpine.DEB.2.21.1907102054400.1758@nanos.tec.linutronix.de> (raw)
In-Reply-To: <27E3B830FA35C7429A77DAEEDEB7344782A9DDD3@IRSMSX103.ger.corp.intel.com>

[-- Attachment #1: Type: text/plain, Size: 2130 bytes --]

On Wed, 10 Jul 2019, Zavras, Alexios wrote:
> J Lovejoy wrote:
> >> On Jul 10, 2019, at 3:38 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> >> On Tue, Jul 09, 2019 at 10:28:59PM -0600, J Lovejoy wrote:
> >>> - the MODULE_LICENSE info was never meant to be definitive license 
> >>> info, but seemingly more of an approximation.  I’m wondering if 
> >>> others have a different view?
> [...]
> >> MODULE_LICENSE predated SPDX by a decade or so, and was designed to 
> >> solve a totally different use case.  I would not try to mix the two, 
> >> or infer one from the other.
> [...]
> > yes. And I can understand the different use case, I guess my concern/question
> > is does the existence of MODULE_LICENSE info that sort of contradicts
> > the actual license info for the file (when looking just at that file,
> > not the combined/resulting image) frustrate the goal of having clean licensing
> > info for when people run scans over the kernel?
> > 
> > or maybe the answer is yes, in a strict scanning sense, but because
> > MODULE_LICENSE is used for a different purpose, so be it… scanners
> > are going to pick it up and people will just have to understand the above?
> > 
> > mostly, I want to confirm that the SPDX identifier for a file in this case
> > can simply be: ISC (not BSD, or GPL) 
> 
> I think we should all agree that MODULE_LICENSE was never intended
> to actually record the exact license -- only to give some information
> on the "GPL-ness" of a module. The naming is unfortunate, but that's
> historical decisions for you and we have to live with it.
> 
> Scanning tools should not rely on it (and its limited set of possible values)
> to determine actual licenses of modules not integrated in the kernel  --
> and an SPDX line with "ISC" (or whatever) should always be the determining input.
> 
> Do we need to document this anywhere to be absolutely clear?

The kernel Documentation of license rules, which also defines the usage of
SPDX identifiers has a section about Module License:

  https://www.kernel.org/doc/html/latest/process/license-rules.html#id1

Is that clear enough?

Thanks,

	tglx